A SIMULATION STUDY OF HYPOTHESIS TESTS FOR DIFFERENCES IN EFFICIENCIES

This paper reports the results of a simulation study to evaluate the p erformance of Banker's (1993) new asymptotic DEA tests for inefficienc y differences between two groups. Their performance is evaluated relat ive to the performance of conventional parametric corrected ordinary l east squares (COLS) tests and the Welch and Mann-Whitney tests used ea rlier in the DEA literature. We consider five different underlying pro duction functions, four different inefficiency distributions and four different sample sizes in our study. Our results include the following : (1) The new asymptotic DEA tests outperform COLS-based tests even wh en the parametric functional form employed for COLS estimation is iden tical to the underlying production function. (2) The asymptotic DEA te sts are robust in that they perform well for different underlying prod uction functions and inefficiency distributions. (3) Welch's mean and Mann-Whitney tests perform worse than both the new asymptotic DEA test s and the COLS-based tests. (4) The performance of COLS-based tests is improved when Banker's (1993) new test statistics are employed using COLS estimates of production inefficiencies. (5) No marked differences in Type II error are found between the various tests with one-output, two-input production technologies. But, the Welch's mean and Mann-Whi tney tests and COLS-based tests perform slightly better than the asymp totic DEA tests in terms of proportion of Type II errors with one-outp ut, five-input production technologies.